With day to day computing becoming difficult. Lot of time and energy is lost in traversing through vehicular traffic. In this paper we propose a solution to predict vehicular traffic using machine learning techniques. US traffic 2015 dataset contains daily volumes of traffic, since this is an actual data and not a simulated one, patterns found in this research can be benefited and made valuable with differing datasets over time. Data driven solutions are proven to breakthrough tough problem statements. Thus in this paper we propose stage by stage machine learning processes to build an efficient model capable of predicting traffic volume based on features which brings out hidden insights in vehicular movements. Inspired by existing research exploration over traffic, in this paper we share our contribution of making traffic predictable thus saving time and energy in day to day life. © 2020 IEEE.